Put the `runId` to the execution plan

[zacaydcloudera]

Hi @wajda @cerveada
can you change the spline code that it can put also the run id to the execution plan?
1.I mean that in the json of execution plan- need to put the JobID
or maybe write a seperate Scala job that can call a method of the agent and put the runID on the executionplan
2.Need to put to hdfs or local system centralize place where the filename will be the ID of the execution plan?
thanks alot

[wajda]

sorry for a slow response. I'm currently very busy with other things. I'll try to go through your messages over the weekend.

Regarding your questions:

1. Assuming Run Id to be the same as Application Id, it's already captured, not in the plan but in the event object. See event.extra.appId
2. Will address it in your another ticket.

[zacaydcloudera]

Can we change the name of the ArangoDB Database as a parameter on the yaml?
if yes- what is the effort for that?

[wajda]

The database name is set on the Spline server. You can make whatever, of course.

See https://absaoss.github.io/spline/#start-spline-server:~:text=SPLINE_DATABASE_CONNECTION_URL

[zacaydcloudera]

1.What about the change of the HDFS to a centralized location? can It be done? and the name of the file to be the as the exectionid+datetime
2.About the name of arangoDB- What i want is the on the spark config... the question has moved here

[wajda]

1. Will answer in the other ticket
2. Not possible. Spline server is single-tenant. You have to have a separate Spline server per customer to achieve your use-case.

[zacaydcloudera]

Thanks
please tell me when you answer No.1

[wajda]

I moved the multi-tenancy topic to the discussions (AbsaOSS/spline#1400) and I'm removing it from here respectively.

[zacaydcloudera]

@wajda Can you answer regarding What about the change of the HDFS to a centralized location? can It be done? and the name of the file to be the as the exectionid+datetime

[wajda]

I will answer when I have time for it. Sorry.